Sunspots and Sea Surface Temperature

Guest Post by Willis Eschenbach

I thought I was done with sunspots … but as the well-known climate scientist Michael Corleone once remarked, “Just when I thought I was out … they pull me back in”.  In this case Marcel Crok, the well-known Dutch climate writer, asked me if I’d seen the paper from Nir Shaviv called “Using the Oceans as a Calorimeter to Quantify the Solar Radiative Forcing”, available here. Dr. Shaviv’s paper claims that both the ocean heat content and the ocean sea surface temperature (SST) vary in step with the ~11 year solar cycle. Although it’s not clear what “we” means when he uses it, he says: thumb its the sun“We find that the total radiative forcing associated with solar cycles variations is about 5 to 7 times larger than just those associated with the TSI variations, thus implying the necessary existence of an amplification mechanism, though without pointing to which one.” Since the ocean heat content data is both spotty and incomplete, I looked to see if the much more extensive SST data actually showed signs of the claimed solar-related variation.

To start with, here’s what Shaviv2008 says about the treatment of the data:

Before deriving the global heat flux from the observed ocean heat content, it is worth while to study in more detail the different data sets we used, and in particular, to better understand their limitations. Since we wish to compare them to each other, we begin by creating comparable data sets, with the same resolution and time range. Thus, we down sample higher resolution data into one year bins and truncate all data sets to the range of 1955 to 2003.

I assume the 1955 start of their data is because the ocean heat content data starts in 1955. Their study uses the HadISST dataset, the “Ice and Sea Surface Temperature” data, so I went to the marvelous KNMI site and got that data to compare to the sunspot data. Here are the untruncated versions of the SIDC sunspot and the HadISST sea surface temperature data.

sidt sunspots and HadISST sea surface temperature 1870 2013Figure 1. Sunspot numbers (upper panel) and sea surface temperatures (lower panel).

So … is there a solar component to the SST data? Well, looking at Figure 1, for starters we can say that if there is a solar component to SST, it’s pretty small. How small? Well, for that we need the math. I often start with a cross-correlation. A cross-correlation looks not only at how well correlated two datasets might be. It also shows how well correlated the two datasets are with a lag between the two. Figure 2 shows the cross-correlation between the sunspots and the SST:

cross correlation sidc sunspots hadISST 1870 2013Figure 2. Cross-correlation, sunspots and sea surface temperatures. Note that they are not significant at any lag, and that’s without accounting for autocorrelation.

So … I’m not seeing anything significant in the cross-correlation over full overlap of the two datasets, which is the period 1870-2013. However, they haven’t used the full dataset, only the part from 1955 to 2003. That’s only 49 years … and right then I start getting nervous. Remember, we’re looking for an 11-year cycle. So results from that particular half-century of data only represent three complete solar cycles, and that’s skinny … but in any case, here’s cross-correlation on the truncated datasets 1955-2003:

cross correlation sidc sunspots hadISST 1955 2003Figure 3. Cross-correlation, truncated sunspots and sea surface temperatures 1955-2003. Note that while they are larger than for the full dataset, they are still not significant at any lag, and that’s without accounting for autocorrelation.

Well, I can see how if all you looked at was the shortened datasets you might believe that there is a correlation between SST and sunspots. Figure 3 at least shows a positive correlation with no lag, one which is almost statistically significant if you ignore autocorrelation.

But remember, in the cross-correlation of the complete dataset shown back in Figure 2, the no-lag correlation is … well … zero. The apparent correlation shown in the half-century dataset disappears entirely when we look at the full 140-year dataset.

This highlights a huge recurring problem with analyzing natural datasets and looking for regular cycles. Regular cycles which are apparently real appear, last for a half century or even a century, and then disappear for a century …

Now, in Shaviv2008, the author suggests a way around this conundrum, viz:

Another way of visualizing the results, is to fold the data over the 11-year solar cycle and average. This reduces the relative contribution of sources uncorrelated with the solar activity as they will tend to average out (whether they are real or noise).

In support of this claim, he shows the following figure:

Shaviv Figure 5Figure 4. This shows Figure 5 from the Shaviv2008 paper. Of interest to this post is the top panel, showing the ostensible variation in the averaged cycles.

Now, I’ve used this technique myself. However, if I were to do it, I wouldn’t do it the way he has. He has aligned the solar minimum at time t=0, and then averaged the data for the 11 years after that. If I were doing it, I think I’d align them at the peak, and then take the averages for say six years on either side of the peak.

But in any case, rather than do it my way, I figured I’d see if I could emulate his results. Unfortunately, I ran into some issues right away when I started to do the actual calculations. Here’s the first issue:

sidc sunspots hadISST 1955 2003Figure 5. The data used in Shaviv2008 to show the putative sunspot-SST relationship.

I’m sure you can see the problem. Because the dataset is so short (n = 49 years), there are only four solar minima—1964, 1976, 1986, and 1996. And since the truncated data ends in 2003, that means that we only have three complete solar cycles during the period.

This leads directly to a second problem, which is the size of the uncertainty of the results of the “folded” data. With only three full cycles to analyze, the uncertainty gets quite large. Here are the three folded datasets, along with the mean and the 95% confidence interval on the mean.

sst anomaly folded over solar cycle 1955-2003Figure 6. Sea surface temperatures from three full solar cycles, “folded” over the 11-year solar cycle as described in Shaviv2008

Now, when I’m looking for a repetitive cycle, I look at the 95% confidence interval of the mean. If the 95%CI includes the zero line, it means the variation is not significant. The problem in Figure 6, of course, is the fact that there are only three cycles in the dataset. As a result, the 95%CI goes “from the floor to the ceiling”, as the saying goes, and the results are not significant in the slightest.

So why does the Shaviv2008 result shown in Figure 4 look so convincing? Well … it’s because he’s only showing one standard error as the uncertainty in his results, when what is relevant is the 95%CI. If he showed the 95%CI, it would be obvious that the results are not significant.

However, none of that matters. Why not? Well, because the claimed effect disappears when we use the full SST and sunspot datasets. Their common period goes from 1870 through 2013, so there are many more cycles to average. Figure 7 shows the same type of “folded” analysis, except this time for the full period 1870-2013:

full sst anomaly folded over solar cycle 1955-2003Figure 7. Sea surface temperatures from all solar cycles from 1870-2013, “folded” over the 11-year solar cycle as described in Shaviv2008

Here, we see the same thing that was revealed by the cross-correlation. The apparent cycle that seemed to be present in the most recent half-century of the data, the apparent cycle that is shown in Shaviv2008, that cycle disappears entirely when we look at the full dataset. And despite having a much narrower 95%CI because we have more data, once again there is no statistically significant departure from zero. At no time do we see anything unexplainable or unusual at all

And so once again, I find that the claims of a connection between the sun and climate evaporate when they are examined closely.

Let me be clear about what I am saying and not saying here. I am NOT saying that the sun doesn’t affect the climate.

What I am saying is that I still haven’t found any convincing sign of the ~11-year sunspot cycle in any climate dataset, nor has anyone pointed out such a dataset. And without that, it’s very hard to believe that even smaller secular variations in solar strength can have a significant effect on the climate.

So, for what I hope will be the final time, let me put out the challenge once again. Where is the climate dataset that shows the ~11-year sunspot/magnetism/cosmic rays/solar wind cycle? Shaviv echoes many others when he claims that there is some unknown amplification mechanism that makes the effects “about 5 to 7 times larger than just those associated with the TSI variations” … however, I’m not seeing it. So where can we find this mystery ~11-year cycle?

Please use whatever kind of analysis you prefer to demonstrate the putative 11-year cycle—”folded” analysis as above, cross-correlation, wavelet analysis, whatever.



My Usual Request: If you disagree with someone, myself included, please QUOTE THE EXACT WORDS YOU DISAGREE WITH. This prevents many flavors of misunderstanding, and lets us all see just what it is that you think is incorrect.

Subject: This post is about the quest for the 11-year solar cycle. It is not about your pet theory about 19.8 year Jupiter/Saturn synoptic cycles. If you wish to write about them, this is not the place. Take it to Tallbloke’s Talkshop, they enjoy discussing those kinds of cycles. Here, I’m looking for the 11-year sunspot cycles in weather data, so let me ask you kindly to restrict your comments to subjects involving those cycles.

Data and Code: I’ve put the sunspot and HadISST annual data online, along with the R computer code, in a single zipped folder called “Shaviv


newest oldest most voted
Notify of

Mother Nature isn’t a joke nor is she easy to comprehend and understand 🙂

Santa Baby

UV is also hot?

Willis says
Where is the climate dataset that shows the ~11-year sunspot/magnetism/cosmic rays/solar wind cycle? Shaviv echoes many others when he claims that there is some unknown amplification mechanism that makes the effects “about 5 to 7 times larger than just those associated with the TSI variations” … however, I’m not seeing it. So where can we find this mystery ~11-year cycle?
henry says
it is very simple really. you must just look at the right parameters. As stated before, I would not look at SSN for various reasons. I am not going to argue that again.
Here are my latest results for the drop in maximum temperatures
change/decrease in maximum temperatures (henry’s global average, 27 stations NH and 27 station SH)
last 40 years (from 1974) +0.034 degree C/yr
last 34 years (from 1980) +0.026 degree C/ yr
last 24 years (from 1990) +0.014 degree C/yr
Here is a graph showing the drop in the magnetic fields from the sun
clearly you can draw a binomial from top and bottom, as a best fit for the general drop in field strength, coming to a lowest point soon? Clearly you can see a binomial for the drop in maximum temperatures?
There must be a correlation between the drop in energy coming in (maxima) and drop in field strength.
(my proposed mechanism) The mechanism is that lower field strengths on the sun allow somewhat more of the most energetic particles to escape from the sun, hence the noted increase in ozone and others TOA. In turn, more ozone and others deflect more sunlight to outer space due to absorbance and re-radiation. Hence we are cooling, globally.


Looking for evidence of solar influence on climate in 11 year cycles is a dead end. Solar variation can only have caused about 0.8C in 150 years. Claiming that because 11 year cycles cannot be seen in noisy incomplete data so solar variation has little influence is also a dead end.
The mechanism for solar influence is simple. The UV frequencies that vary most greatly between cycles are the frequencies that penetrate below the diurnal overturning layer of the oceans, allowing energy to accumulate. This process is slow and cumulative. The apparent absence of clear 11 year cycles in ocean temps in no way shape or form disproves solar variation driving climate changes observed over 150 years.
In contrast, AGW due to CO2 is easily disproved. Empirical experiment shows the oceans are not warmed by DWLWIR. Empirical experiment also shows that the oceans respond as a selective surface to incoming solar radiation not as a “near blackbody” as per the crazed claims of climastrologists. Further, the effective emissivity (not the apparent emissivity) of water is below 0.8. The oceans need the atmosphere to cool and the atmosphere in turn needs radiative gases to cool. Global warming due to CO2 is physically impossible.
That only leaves two options for 0.8C in 150 years. Internal variability or solar variability. Solar variability is prime suspect. All you need to understand is that water is not a “near blackbody” or even close.

Frederick Michael

Shouldn’t the sunspots (or whatever the real solar driver is) drive the FIRST DERIVATIVE of temperature, not simply the temperature? If so, then the earth’s temperature might have a slow impulse response which could be much longer than 11 years. The result could easily act like a low-pass filter — and you wouldn’t see much happening within the 11 year periodicity of the sunspot cycles. BUT, give us have a few weak cycles in a row and the accumulated effect would be significant.
Given the radiative heat loss that results from higher temperatures, a good model (of global temp as a function of sunspots) might be some kind of exponential smoothing. That would be true even if global temp responded quickly enough to “follow” the 11 year solar cycle, but I’m thinking that the right smoothing constant would be too small for that.
Another way to think of this is as a low-pass RC circuit. The input is a current source which varies over time (but is not AC; it’s always >0). The circuit is a capacitor to ground (representing the thermal inertial of the earth) and a resistor to ground (representing the radiative loss as a function of temperature.) Over a small temperature range, we can model the radiative loss as proportional to temperature and ignore the higher order effects.
Either way, it’d be interesting to see if you can get a better fit, showing temperature rising after strong solar cycles and falling after weaker ones. Unfortunately, the El Nino/La Nina cycles add a hell of a lot of noise to the data.

So, let’s play the ‘maybe sunspots-co-relate-to-something-else-in-the-sun-that-might-affect-global temperatures on earth over long periods of time game. It does make sense somehow, but we may not (actually really and absolutely do not know what) that relationship might be.
Look again above at the plots.
Add a lag.
Look not a 11 year cycle, but a six-sunspot 11 year (33 positive/33 year negative) 66 year cycle of alternating “positive” and “negative” cycles that themselves are near-equal, but over a three set cycle may mean something important .


These studies by Willis aimed at reproducing important (apparently) research findings are of enormous scientific value. The issue of research repeatability has been highlighted recently in this Nature editorial concerning preclinical cancer research:
In short, the pharmaceutical company Amgen in California tried to repeat 53 “landmark” cancer genetic studies and was able to do so in only 6 cases. The German company Bayer tried the same thing with a different (partly overlapping) set of studies and could reproduce only 25% of them.
This shocking result is prompting a change in research and publication practice with more encouragement of attemps to repeat published research. While not as glamourous as blazing a trail with an (apparently) original finding, it does a service to science of great value.

Look again above at the plots.
Henry says
Look at this one
Hale cycle (the “plus” + the “minus” ) is from 1970-1990 and the next is from 1990-2013.
However, fleld strengths are so low that it must come to a dead end stop, probably reversing. My bet is on 2015 or 2016. For the next 44 years (from 2015 or 2016) field strengths will be the mirror of previous 44 years, unless somebody knows better?

richard verney

Konrad says:
June 6, 2014 at 11:51 pm
We hold similar views, and recall that we engaged in much discussion on this on Willis’ article on ‘Radiating the Oceans’ (which i personally consider to be not one of Willis’ stronger articles – sorry willis, just my personal opinion). However, the point that the warmists would raise is that if DWLWIR heats the atmosphere, even if it does not heat the ocean below, then due to the warmer atmosphere above the ocean,the heat loss from the ocean is lower/slower, thereby helping to maintain or even produce higher ocean temperatures over time.
In another Willis’ article (I can’t rember which one), I pointed out that if one considers the optical absorption characteristics of LWIR in water, if DWLWIR is of the level claimed by K&T (in their energy budget cartoon), and if that energy is absorbed by the oceans in accordance with accepted LWIR absorption charactericis in water, there would be so much energy absorbed in the first few micron layer that this potentially would produce about 14 to 18 metres of rainful annually. I suggested that since we are not observing such levels of annual rainfall, it suggests that DWLWIR is not of the level claimed, or it lacks sensible energy, or it is simply not (for some other reason) being absorbed by the oceans.
Indeed, I am sceptical as to whether all the DWLWIR can even reach the oceans. The average wind conditions over the oceans s BF 4 to 5. in these wind conditions, there is already quite a lot of wind swept spray and spume. Spray and spume is, of course, a fine mist of water droplets, and these droplets are more than a few microns in size and therefore would be capable of absorbing about 60% of all DWLWIR before it even reaches the ocean below.
Now I am not suggesting that there is a homogenous layer of windswept spray and spume which universally covers all the oceans all of the time. But what one has to rememeber is that globally, as you read this there are large areas of the oceans which are subject to BF 7 to 8 and above. Indeed, there will be areas of the ocean where large storms are raging with BF10 conditions. We have all seen the size of huricanes and cyclones, and one can imagine the area and sea state involved. Where these storms are raging, the spray and spume which is a layer of water droplets completely divorced from the ocean below, acts much like sun cream (or a sun parasol), but blocking DWLWIR from impacting the ocean below. So it would appear that a not insignificant proportion of the DWLWIR in the K&T energy budget cartoon, does not, or cannot reach the ocean below.
The divorced layer of spray and spume in these conditions would almost fully absorb DWLWIR and as it does so, it would heat and would be carried upwards in the atmosphere initally warming the atmosphere and keeping the DWLWIR away from the ocean below.
I have yet to see a wholly convincing argument as to why if DWLWIR is of the order suggested by K&T (in their energy budget cartoon), and if DWLWIR is absorpbed by the oceans in accordance with accepted LWIR absorption charateristics in water, there is not a very substantial amount of rainfall (much more than we observe annually) and/or that this would not lead to copious amounts of evaporation.
When considering this, one has to bear in mind that the heat flux is upwards in the first few millimetres of the ocean such that energy absorbed in the first few microns cannot find its way downwards by conduction, and ocean over turning is a slow mechanical process such that even if the top few microns of the ocean are over turned, the rate of overturning would be slower than the speed at which energy is absorption in the first few micron layer 9such that the ocean over turning process does not disipate energy downwards at a quick enough rate to stop the rapid evaporation that one would expect to see at the top of the ocean give the amount and rate of absorption of DWLWIR in the first 3 or 4 microns of the ocean).

HenryP says:
June 7, 2014 at 12:38 am (replying to) RACookPE1978

Henry says
Look at this one
Hale cycle (the “plus” + the “minus” ) is from 1970-1990 and the next is from 1990-2013.

Oh, I can – And will!! – agree with you about the coming slower sunspot/solar cycle-magnetic field total questions.
But!!!! I do NOT know what will happen due to that change.
So, to cover for that lack of knowledge, I would prefer to focus on the earlier longer-term 66 year patterns of “several high, several low” cycles we see since 1650 as the earth warms from the LIA. Do those cycles matter?

Peter Hartley

A problem with using a long time series of SST is that early measures are not comparable to more recent ones. There has been a long debate about adjustments for different buckets used to measure sea water temperatures, buckets versus engine intake, and the changing coverage of the oceans because temperatures were only measured where ships went in the pre-satellite/ARGO days. I seem to recall that both John Daly and Steve McIntyre have discussed these measurement issues.

Konrad says
Solar variability is prime suspect.
Henry says
according to AGW theory (warming caused by more CO2) minimum temperatures should show a rise. Namely, it is alleged that increased GHG causes a delay in cooling.
Consequently, minimum temps. should be rising.
Here are my latest results for the change in the speed of minimum temperatures (27 weather stations NH + 27 weather stations NH, balanced to zero latitude and 70/30 @sea/inland)
last 40 years (from 1974) +0.004 degree C/yr
last 34 years (from 1980) +0.007 degree C/ yr
last 24 years (from 1990) +0.004 degree C/yr
last 14 years (from 2000) -0.009 degree/yr
Now, note that the observed values are very low, indeed, yet it seems they are significant.
Namely, setting the periods out against the speed of warming/cooling I get a binomial again with rsquared eual to 1 (100% correlation)
There is no error in the equation…..
Hence, there is no AGW There is no room for it in my equation.
unless AGW behaves naturally?
Temperature depends on solar variability only.
Have a good weekend.

Willis, I did find a graph from NASA used for teachers which discusses sunspots at:
At least one can say that the correlation is weak. But another more recent work by Zhou and Tung shows a correlation:
Another article I have read was that the SST in the (sub)tropics rapidely increases with 0.3-0.5 K and back over a cycle, but I lost the link.
I am agnostic on that point (solar cycle influence on SST), but there is definitely a response of weather patterns on the solar cycle via the UV-route: UV increases about 10% during solar maxima, which alters O3 abundance in the tropical lower stratosphere, increases its temperature (1 K), increases the temperature difference with the poles at that height and shifts the jetstreams polewards, including the connected wind-, cloud- and rainpatterns. That is reflected in river flows. Several findings show the correlations: (full article?) (rivers in Portugal) (Mississippi catch area) (Mississippi / Missouri)
similar correlations were found for the Nile (Egypt), Po (Italy) and South African rivers, but the URL’s I had don’t work anymore…
The main article that shows the UV-jetstream connection is not on line anymore, but I have several others which connect solar cycles, stratospheric influences and weather/climate on earth:
A lot of stuff to analyse for you…

Greg Goodman

Interesting work as always Willis.
Cross-correlation is a good start. I suggest you run some kind of FT on that. Just by eye I’d say that in fig2 you have a strong component about 11y mixed with something longer >20y

Jimmy Haigh

How does the solar cycle correlate with the orbit of Jupiter (11.86yrs)? Or perhaps the combined effect of Saturn and Jupiter which come together every 22 years or so?

Henry says
Henry says again
so sorry, that one sentence should read
27 weather stations NH + 27 weather stations SH, balanced to zero latitude and 70/30 @sea/inland)


“Jimmy Haigh says:
June 7, 2014 at 1:07 am ”
How dare you! We know, in computer simulations, that only CO2 drives climate on this rock and only that esitmated ~3% of ~390ppm/v at that from driving cars and using lights etc.
I think a /sarc off tag is not require here.

richard verney

At the crux of the point raised by Willis is that all the various data sets that are used in climate science are not fit for purpose. Unfortunately, this is not sufficiently recognised and/or accepted by the scientist who seek to use those data sets.
Unfortunately, they are all either of too short a duration, and/or have too wide an error margin and/or not enough sample saturation and/or are horribly bastardised by dropouts, pollution by UHI and/or endless bastardisation caused by adjustments made to the data set, the need for and correctness of which is moot.
This is one reason why one cannot see any first order correlation between the rise of CO2 levels and temperature in any of the instrument data sets. The signal from CO2 (if indeed there is any signal) is too small to be revealed in the noisy data sets that we have available, especially given the margin of error in the measurements undertaken.
If there is indeed manmade global warming, the only two data sets of note are the CO2 data set and ocean temperture data sets (ocean temperatures are a metric for energy absorption and hence imbalance, and the heat capacicty and energy stored in the oceans overwhelms that of the atmosphere by orders of magnitude).
Unfortunately, pre ARGO, there is no reliable ocean temperature data, and even post ARGO there are problems; first the adjustment made to ARGO data when it came on stream which data suggested that the oceans were cooling, second, the coverage – the oceans are vast and there are few ARGO buoys given the area and volume involved, and third, it is not known whether there is an in built bias caused by the free floating nature of the buoys which may have a tenedancy to be influenced by ocean currents which currents are themselves an artifact of variations in ocean temperature and/or density).
I am not surprised by Willis’ evaluation. It is what one would expect given the inadequacies of the available data sets, and the tendancy for scientists to over stretch the limits of the data available. Further the very nature of a coupled non linear chaotic system makes identification of trends and signals and corresponding responses extremely difficult to detect.

Charles Nelson

The ‘Global’ Sea Surface temperature anomaly in 1870 was positive ‘0.2C’
Let me just get this straight, you’re talking about the ‘year’ 1870 and you’re saying that somebody measured the ‘global’ sea surface temperature to within 1/10 of a degree C back then and came up with the 0.2C anomaly?

Greg Goodman

The presence of a longer term periodic change would also explain why the 11y “folding” (which is a pretty inappropriate term since it implies some kind of reversal) ends up with just noise. If the long period is near 22y using that as the offset may be more appropriate.
Frederick Michael says:
“Shouldn’t the sunspots (or whatever the real solar driver is) drive the FIRST DERIVATIVE of temperature, not simply the temperature? ”
Indeed. The primary effect of a radiative forcing is dT/dt as can be seen by the physical dimensions. Radiation flux is power ; temp is energy. They are orthogonal, so the initial investigation should be rad (SSN) vs temp.
If there is an effect it shoud accumulate but integrating with the huge capacity of the oceans will great smooth out any signal to the point where Willis’ one-size-fits-all 0.2 significance test will fail.
There is auto-correlation in temperature for precisely this reason. Willis makes reference to it on two occassions but seems imply that the results will be even worse if he accounted for autocorrelation, without saying why.
The most current way to remove autocorrelation is by taking the first different. This would in fact be dT/dt !

richard verney

Just a short point which is applicable to nearly all considerations regarding the effects of Solar irradiance, one cannot properly consider this in the absence of reliable data on cloud cover.
Until we know the extent of cloudiness over time (area, volume, composition, height of cloud stack, time of formation, duration of formation, place of formation, the underlying albedo which is being shielded by cloud cover, the surface type below the cloud cover and its absorption characteristics etc), one cannot reasonably consider how much energy is being imparted to the surface and without that one cannot begin to consider what sort of response one is expecting to see.


“Charles Nelson says:
June 7, 2014 at 1:16 am”
That’s rather inconvenient. NOAA say they “know” what the global average sea and land temperatures were in 1880. I dunno, sounds like it was made up to me.

Santa Baby

“That only leaves two options for 0.8C in 150 years. Internal variability or solar variability. Solar variability is prime suspect. All you need to understand is that water is not a “near blackbody” or even close.”
There has been to much UHI and political motivated correction of the data sets. First one will have to remove these before one can use the data to find a correlation with anything.
Before the previous. IPCC report the Jones et al showed the arctic North of 70 deg North to have been as warm or warmer in the 1930s. That changed drastically overnight with a new data set.
If arctic really is now warmer than in the 1930s, then I find it strange that the global temperature is much higher than in the 1930s. UHI and political motivated corrections?

‘Although it’s not clear what “we” means when he uses it’
That’s the Royal “We”. We use it ourselves, in our sole-authored papers.

Greg Goodman
cross-correlation of monthly global SST and SIDC monthly SSN.
As I suspected a notable circa 22y peak and “11y” is split into fine structure as is pretty much universally found, this is not a single peak. Closer read-off here gives : 10.12 and 11.22 y.
The largest peak is at 170y and is ten times the magnitude of the peaks show in this detail.
Usual caveats about data length etc apply. But that was the long period found in SSN in the chinese paper featured yesterday on WUWT. It appears also in cross-correlation with global SST as the strongest signal.
I’m curious that peaks appear to be close to being multiples 5.5 11,22,33,44,66


“….smaller secular variations in solar strength can have a significant effect on the climate…”
I have a problem with the words ‘smaller’ and ‘significant’ …… ‘smaller’ is the TSI variation, because of the huge output of the Sun….. the effect ‘significant’ only for humans who want to be comfy in a ‘narrow’ range of climate…. at the moment global warming is ~ 0.5deg extra….. well 300K or 300.5K ….. bugger all difference …. but ‘significant’ for humans….

The correlation between Earth Sea & Land Temps is from 22 year Magnetic Cycles NOT 11 year periods, why do you insist to show the 11 year Solar Cycles when it doesn’t show/prove anything as you have now demonstrated. Wills please go back to the drawing board and re work this article, I’m sure your find what your looking for

David McKeever

myline=strsplit(discrets(as.vector(spotgauss),collapse=T),split=””);myline (line 43 of shavivcorrelations.R) made my RStudio squawk. I couldn’t see a discrets() or discretes() function in your Willis Functions, and a search showed some possibles discretes() in the ggplot package.


here is clear the magnetic activity of the Sun and the Earth’s magnetic field.


Have you looked at this ?
Graph below is a detrended historical plot of the sea surface temperature anomalies (HADSST3) for the Pacific and Atlantic Ocean basins from pole to pole The peaks and valleys of this plot match the peaks and valleys of global atmospheric cooling and warming periods over the last 130 years . The surface temperatures of these oceans have peaked and are again heading for a cold trough by about 2040/2045 like they did 1910 and 1975. A global warming peak like we recently had is not predicted for 65- 70 years or until 2075/2080. The source of this cooling is the Global oceans Meridonal Overturning Circulation or MOC When there is a stronger than normal MOC, there is more deep cold water upwelling into the oceans by means of ocean conveyor belts. This will ultimately cool the SST and cool the Arctic as is already happening.
Courtesy of Bob Tisdale’s and WUWT web pages
If we hind cast the above ocean graph and in particular the 70 year Atlantic Ocean SST, pole to pole , we find that major SST troughs like 1905/1910 and 1975 could have also happened in 1835, 1765, 1695 and major peaks in SST like 2010 and 1940 could have happened in 1870, 1800, 1730 and 1660.
For example the North Atlantic Ocean may have been cooling during the following past periods [And probably the Pacific as well.] The major solar minimum period is also noted
1940 to 1975
1870 to 1910[Minimum 1880-1910]
1800 to 1835[Dalton minimum 1790-1820]
1730 to 1765
1660 to 1695 [Maunder minimum 1645-1715]
1590 to 1625
1520 to 1555 [Sporer minimum 1460-1550]
1450 to 1485 [ Sporer minimum 1460-1550]
These cooler Atlantic Ocean SST periods correspond to the historic low CET temperatures and just happen to occur during the Maunder, Dalton and Modern Minimums of 1645-1715, 1790- 1820, and 1880-1910. In another words the reason for the low CET temperatures could have been the cool Atlantic SST and not because of the changing solar cycle during each of the three major solar minimums.
This changing Atlantic Ocean pattern can be seen in this Reconstructed North Atlantic SST between 1567 and 1990 with the courtesy of Bob Tisdale’s web page
Courtesy of Bob Tisdale


Now let’s look at the ocean temperature anomalies.

My impression is that these types of studies are excellent examples of what is known as “…torturing and molesting…” the data.

RACookPE1978 says
But!!!! I do NOT know what will happen due to that change.
So, to cover for that lack of knowledge, I would prefer to focus on the earlier longer-term 66 year patterns of “several high, several low” cycles we see since 1650 as the earth warms from the LIA. Do those cycles matter?
henry says
we can measure the change if we take a balanced sample of weather stations of the world as shown earlier up this thread, and determine the change in temperature per annum
(which Greg also refers to)
Here are my latest results for the change in the speed of minimum temperatures (27 weather stations NH + 27 weather stations SH, balanced to zero latitude and 70/30 @sea/inland)
last 40 years (from 1974) +0.004 degree C/yr
last 34 years (from 1980) +0.007 degree C/ yr
last 24 years (from 1990) +0.004 degree C/yr
last 14 years (from 2000) -0.009 degree/yr
Now, note that the observed values are very low, indeed, yet it seems they are significant.
Namely, setting the periods out against the speed of warming/cooling I get a binomial again with rsquared equal to 1 (100% correlation)
There is no error in the equation…..
Hence, there is no AGW… There is no room for it in my equation.
One is tempted to think that we can project on this binomial forward, which would imply more cooling coming up ahead. Indeed, I think some more cooling is still coming up ahead. But we know from various investigations of mine, including ozone increase and the evaluation of solar magnetic field strengths, that we must come to a dead end stop on that binomial. Everything points to 2015 or 2016 as the date for the dead end stop of the process that causes the cooling on earth (coming from a variance of the sun’s output)
which suggests a 87 or 88 – and a 208 year cycle when we look at energy coming in
e.g. also
Persistence of the Gleissberg 88-year solar cycle over the last ˜12,000 years: Evidence from cosmogenic isotopes
Peristykh, Alexei N.; Damon, Paul E.
Journal of Geophysical Research (Space Physics), Volume 108, Issue A1, pp. SSH 1-1, CiteID 1003, DOI 10.1029/2002JA009390
Among other longer-than-22-year periods in Fourier spectra of various solar-terrestrial records, the 88-year cycle is unique, because it can be directly linked to the cyclic activity of sunspot formation. Variations of amplitude as well as of period of the Schwabe 11-year cycle of sunspot activity have actually been known for a long time and a ca. 80-year cycle was detected in those variations. Manifestations of such secular periodic processes were reported in a broad variety of solar, solar-terrestrial, and terrestrial climatic phenomena. Confirmation of the existence of the Gleissberg cycle in long solar-terrestrial records as well as the question of its stability is of great significance for solar dynamo theories. For that perspective, we examined the longest detailed cosmogenic isotope record—INTCAL98 calibration record of atmospheric 14C abundance. The most detailed precisely dated part of the record extends back to ˜11,854 years B.P. During this whole period, the Gleissberg cycle in 14C concentration has a period of 87.8 years and an average amplitude of ˜1‰ (in Δ14C units). Spectral analysis indicates in frequency domain by sidebands of the combination tones at periods of ≈91.5 ± 0.1 and ≈84.6 ± 0.1 years that the amplitude of the Gleissberg cycle appears to be modulated by other long-term quasiperiodic process of timescale ˜2000 years. This is confirmed directly in time domain by bandpass filtering and time-frequency analysis of the record. Also, there is additional evidence in the frequency domain for the modulation of the Gleissberg cycle by other millennial scale processes.
end quote
hope this helps

There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.
Mark Twain
US humorist, novelist, short story author, & wit (1835 – 1910)

Greg Goodman

Here is the correlation function of SST and SSN from which I derived the spectrum I posted above.
There is a rather broad peak in correlation at a lag of about 10 years and an equally strong, but more focused correlation at just over 21 years. This explains why Willis’ extension of the 11y “folding” idea shows nothing of interest. It does disprove either the circa 11 or 22y cycles or the presence of a solar signal. In fact is was a fairly sure way to destroy what is there. ( This was not Willis’ own idea, he was extending a technique used by Nir Shaviv. )
It should also be noted that is a clear anti-correlation with period of about 140 years and a lag of half that. That is a period of time, there is not enough data to suggest this is periodic as in cyclically repetitive.
Even this cursory investigation seems enough to suggest there is a correlation between SST and solar activity represented by SSN.


Willis, I am living on a Sailboat in the Bahamas. The surface temperature as measured by my IR gun for the entire last week was 80.4˚ F. And today since the wind has quit blowing the Surface temperature jumped to 84.6˚ F. The surface temperature is incredibly consistent 24 hours a day as long as the winds are consistent.
The ocean surface temperatures are controlled almost entirely by windspeed (evaporation rate). Clouds, Sunlight, clear sky, and night have no effect on the surface temperature.
Sure there are other things that affect the surface temperature, length of day, currents, rain, barometric pressure, but they are minor. The dominant factor is the wind.
If you think about it, this observation explains the UHI, Global warming and disproves climate change.

An IR gun. Calibrated? What’s the emissivity of the sea surface? Cross verified with a good old mercury in glass?

Hi Willis,
Despite being a regular reader here and at other climate blogs, I’ve not before come across the correlation between solar irradiance and poleward energy flux that Botkin recently put up as evidence to the Senate (of solar influence). I assume this is related to the (paywalled) paper below? (all I could find on a quick search). I’m not a climate dude so haven’t the tools to investigate. Have you come across this before and do you have any opinion on the apparent correlation? I can see by eye a couple of obvious deviations, but these could be volcanoes or something. No worries if I’m opening another can of worms you don’t want to look inside of 😉
Solar irradiance modulation of Equator-to-Pole (Arctic) temperature gradients: Empirical evidence for climate variation on multi-decadal timescales: Willie Soon and David R. Legates, 2013.


I just came across this yesterday. From climateaudit:

Update: In response to an inquiry from a CA reader, Holgate gave the following response on solar/sea level connections:
Holgate’s response:
Many people have tried to link climate variations to sunspot cycles. My own feeling is that they both happen to exhibit variability on the same timescales without being causal. No one has yet shown a mechanism you understand. There is also no trend in the sunspot cycle so that can’t explain the overall rise in sea levels even if it could explain the variability. If someone can come up with a mechanism then I’d be open to that possibility but at present it doesn’t look likely to me.
If you’re interested in solar cycles and sea level, you might look at a paper written by my boss a few years back: Woodworth, P.L. “A world-wide search for the 11-yr solar cycle in mean sea-level records.” Geophysical Journal of the Royal Astronomical Society. 80(3) pp743-755
You’ll appreciate that this is a well-trodden path. My own feeling is that it’s not the determining factor in sea level rise, or even accounts for the trend, but there may be something in the variability. I’m just surprised that if there is, it hasn’t been clearly shown yet.

IPCC AR5 Figure 13.6
According to satellite telemetry GMSL between 2005 and 2012.5 increased about 20 mm. That’s about 0.75 of an inch. At that rate the sea level will increase another 8.25 inches by 2100. That won’t be a problem for anybody not dumb enough to currently live or build 4 inches above GMSL. Might need your high water pants.
What sea level rise are you referencing? Something IPCC doesn’t know about?

Nick Yates

Richard Verney says:
When considering this, one has to bear in mind that the heat flux is upwards in the first few millimetres of the ocean such that energy absorbed in the first few microns cannot find its way downwards by conduction
I was just wondering, as water is at it’s densest at 4C, could this make any difference, perhaps nearer the poles?

peter bartner

From a scientist who has read extensively in the field of climate
I would think that the Sun would act mainly through its interactions with our oceans; primarily with UV photons that get by the ozone layer. Visible will also contribute. During the decline of the last solar sunspot cycle, UV radiation declined 6% which should have an impact on ocean temperatures. This mechanism, because of the slow response time of these vast reservoirs, will not respond quickly enough to reflect Sun spot cycles, thus, there would be observable sun spot pattern. .

David L. Hagen

Sea Surface Temperature vs Integral of Total Solar Insolation
Willis. Try comparing Ocean Surface Temperature vs the INTEGRAL of TSI (or sunspots) rather than directly.
Frederick Michael asks: “Shouldn’t the sunspots . . . drive the FIRST DERIVATIVE of temperature, not simply the temperature?”
Greg Goodman affirms: “If there is an effect it should accumulate but integrating with the huge capacity of the oceans will great smooth out any signal”.
See David Stockwell who models and quantitatively shows temperature varies not directly but as the integral of solar flux, with a phase lag of Pi/2 (90 degrees) i.e. 2.75 years lagged from the ~11 year sunspot cycle.
Stockwell shows that the direct correlation of solar irradiance with temperature R^2 is only 0.028 while the cumulative solar irradiance has a correlation R^2 of 0.72 and solar + volcanic has R^2 of 0.78. See Fig. 4 in
David R.B. Stockwell “On the Dynamics of Global Temperature” August 2, 2011
David R.B. Stockwell, “Accumulation of Solar Irradiance Anomaly as a Mechanism for Global Temperature Dynamics” 9 Aug. 2011

Here is presented a novel empirical and physically-based auto-regressive AR(1) model, where temperature response is the integral of the magnitude of solar forcing over its duration, and amplification increases with depth in the atmospheric/ocean system. The model explains 76% of the variation in GT from the 1950s by solar heating at a rate of $0.06\pm 0.03K W^{-1}m^{-2}Yr^{-1}$ relative to the solar constant of $1366Wm^{-2}$.

David R.B. Stockwell, “Key evidence for the accumulative model of high solar influence on global temperature” 4 August 23, 2011

Firstly, variations in global temperature at all time scales are more correlated with the accumulated solar anomaly than with direct solar radiation. Secondly, accumulated solar anomaly and sunspot count fits the global temperature from 1900, including the rapid increase in temperature since 1950, and the flat temperature since the turn of the century. The third, crucial piece of evidence is a 90 deg shift in the phase of the response of temperature to the 11 year solar cycle. These results, together with previous physical justifications, show that the accumulation of solar anomaly is a viable explanation for climate change without recourse to changes in heat-trapping greenhouse gasses.

Stockwell further shows a 2.75 year Phase Shift in Spencer’s Data
Sunspots are an approximate measure of “Total Solar Insolation”.
Richard Verney makes another critically important point in insolation AFTER cloud effects – where there is very sparce uncertain data. David Stockwell highlights the importance of the insolation temperature phase lag to address Spencer’s challenge on whether:

1. Changes in cloud cover actually do drive changes in global temperature due to gamma-ray flux (GRF) or other effects, or
2. The changes in cloud cover are caused by changes in global temperature, with the derivative mechanism described above.
3. Both 1 and 2.

CO2skeptic’s note on the 22 year Hale cycle instead of 11 year Schwab cycle may be significant.


richard verney says:
June 7, 2014 at 12:49 am
You raise valid points regarding ocean spray and near surface saturated air. However while this H2O stops most DWLWIR from the atmosphere reaching the ocean surface, it is itself emitting LWIR back to the surface. There are two questions here, the effect of incident LWIR on the cooling rate of water and the effective emissivity of water.
Empirical experiment proves DWLWIR cannot heat nor slow the cooling rate of water that is free to evaporatively cool. That’s game over for DWLWIR slowing the cooling rate of 71% of Planet Ocean’s surface. You say “sorry Willis”. You wouldn’t say that if you saw what he wrote at Talkshop in 2011. Willis did not lose the debate to me. He lost to a roll of microwave safe cling wrap. That’s just sad.
And the effective emissivity of water? This is one of the church of radiative climastrology’s greatest “mistakes”. I have recently run some IR measurement experiments of warm water under a cryo cooled “sky”. An emissivity setting above 0.95 works well for measuring water temp in a sea of environmental IR. Eliminate this background IR and I find that you need an emissivity setting below 0.8 for accurate reading. I am beginning to suspect the old texts claiming IR emissivity of 0.67 may be correct.
What does this mean?
The ability of water to absorb UV/SW/SWIR is around 0.92. It’s ability to radiate LWIR could be as low as 0.7. Without evaporative cooling our oceans would become a giant evaporation constrained solar pond with temps topping 80C. The atmosphere is provably cooling 71% of the planets surface.
What is the only effective cooling mechanism for the atmosphere?
Radiative gases.
There is only one logical conclusion –
97% of climastrologists are assclowns.

Greg Goodman

David L. Hagen says:
Sea Surface Temperature vs Integral of Total Solar Insolation
It’s essentially the same as dT/dt vs SSN.
differentials amplify the short term signals , integrals smooth them out.
I redid the power spectrum using d/dt(SST) and it’s essentially the same, though I had to filter out the annual variation in SST to resolve the circa 10 and 11y peaks from one another.
One thing that is a surprise is a fairly strong peak at about 9.04 to 9.09 . I also found this while discussing another Willis thread that looked at correlation of sea level and SSN.
I had previously been inclined to think this peak in SST was a lunar tidal effect, but seeing present in these two cases of cross-correlation with SSN suggests it too is solar in origin.
Another pointer to the “fundamental connectedness of all things” to quote Douglas Adam’s Dirk Gently.
This 9.04 is also found in cross-correlation of AMO and PDO (BEST, Curry) and directly in N.Alt and N.Pacific SST. (me).

Greg Goodman

IIRC BEST/Curry paper reported 9.1 +/- 0.4 y , a rather large and cautious uncertainty from Monte Carlo chopping and changing techniques.
I found the peak to be centred on 9.05 in that case ( N.Alt and N.Pacific SST)
This would seem to be the same thing as Scafetta’s 9.01 +/-0.1y

Greg Goodman

Oops. This would seem to be the same thing as Scafetta’s 9.1 +/-0.1y


Someone will say an increase of neutrons of a few percent is not much. Is it really?
Primary cosmic rays are a stream of particles with energies from 10^7 to about 10^20 eV reaching the vicinity of the Earth from interstellar space. It consists of different types of particles: electrons and positrons, protons, alpha particles and heavier nuclei (up to uranium) and gamma quanta with high energy. This radiation does not reach directly to the surface of the Earth and can be seen above the Earth’s atmosphere on satellites or balloons.
The particles of primary cosmic rays entering the Earth’s atmosphere to produce particles called avalanche. secondary cosmic radiation, which is part of the natural radioactivity observed on the surface of the Earth.
Secondary cosmic rays reaching the Earth’s surface consists mainly of muons. Muons are unstable elementary particles very similar to electrons but 200 times heavier. Formed in the atmosphere mainly from the decay of another type of elementary particles, mainly risers and kaons. These in turn are created in the collisions of particles of primary cosmic rays with the atmosphere or the collisions of secondary particles created in the previous collisions.
The muons have an average life of at rest about 2 * 10^-6 seconds (2 microseconds). The muons decay into electrons or positrons and neutrinos. Muons are very penetrating particles (eg if they have energy above 5 billion electron volts, can penetrate more than 10 meters under the ground). With the electric charge can be easily recorded. These are ionizing particles.
The intensity of muons at the surface is about 200 particles per area of ​​1 square meter per second. This corresponds to the passage of particles about 6 per second by the head of a human, resulting in the head ionization approximately 100 million per second. It is natural radiation in our environment!

Jeff L

For those readers with a background in signal processing (especially geophysical processing), please take the time to read this post.
There is a gut feeling among many observers that the sun & solar cycles is somehow connected to climate. A reasonable hypothesis to want to explore , given that the sun ultimately powers our atmosphere.
Willis & others have shown that a solar signal isn’t readily discernible is various atmospheric / oceanic data sets. Perhaps the problem is the assumption of some sort of direct correlation.
Maybe a better mathematical model is a convolutional model ? The time series of some measure of solar activity would be the input signal. This would be convolved with an atmospheric “filter operator” with the output being an observed time series of atmospheric temperature. Since we have both the input signal (solar activity time series) and output series (atmospheric temperature time series), in theory , we could use a deconvolution process to solve for an atmospheric filter response operator.
This operator could then be tested against other datasets for viability in hindcasting and perhaps modified to come up with a better operator with better hindcasting ability. If the process worked, it could establish a connection between solar activity & temperature not readily visible through correlation or cross-correlation or any other spectral based approaches (because the spectra of the output signal would be different than the input signal, filtered by the atmospheric operator).
The spectral characteristics of the atmospheric operator may , in and of itself , provide deep insight into big picture atmospheric processes which have yet been unrecognized. There is an assumption that the operator is stationary but that could also be tested by deriving operators for different time periods. Perhaps there would be predictable variations in the operator with time. that would be useful in making forecasts of future temperature.
In short, there is a ton of potential research that could be done around the convolutional model & there is no telling just what insights could be gleaned until the research is done.
For any geophysicists out there, I am sure you can see the analogy here with seismic data. Input is your seismic source (analog = Solar activity time series) which is convolved with the Earth filter (analog = atmospheric filter ) , resulting in your observed seismic time series trace (analog = observed temperature time series).
I am not sure if anyone has pursued this type of research. I have always wanted to do this, but I simply do not have the time to pursue it. I would love it if someone would pick this idea up & run with it. it may work, it may not but I would be very curious to see the results. If someone chooses to pursue this idea based on this post, please just acknowledge where the idea originated.

Greg Goodman

David L. Hagen says:
Stockwell : “Spencer argues that it is impossible to distinguish between 1 and 2. Both Spencer and Lindzen both consider the lags important because correlation is greatly improved (and determines whether feedback is positive to negative). Neither seem to have mentioned the 3 month phase relationships emerging from integral/derivative system dynamics. I can’t see how it is possible perform a valid analysis without this insight. ”

Greg Goodman

“Maybe a better mathematical model is a convolutional model ? ”
See the above link. Exactly what I did with AOD.

David Riser

I would propose that if you are right, that the temp has a significant regulating mechanism, then any correlation between SSN and any temp set would be very difficult to find as long as there is sufficient energy into the system. Obviously it occasionally gets pretty cold here on earth we have plenty of evidence of that and it gets a tiny bit warmer, there is some evidence of that. So I would propose that we wont have any statistical evidence of sun influence until we enter a period of significant temperature change. .8C just isn’t enough to be detectable. Just my thought. Love your work by the way.
David Riser